基于Curvelet变换与自适应PCNN的红外与可见光图像融合

A Better Algorithm for Fusion of Infrared and Visible Image Based on Curvelet Transform and Adaptive Pulse Coupled Neural Networks(PCNN)

根据红外与可见光图像的成像特点,提出一种基于Curvelet变换与自适应PCNN(Pulse Cou-pled Neural Networks)的图像融合新算法。首先对两幅原始图像进行快速离散Curvelet变换,得到不同尺度与方向下的子带系数;对低频系数采取加权平均融合规则,将高频系数作为PCNN的输入,选取区域能量测度为PCNN的连接强度,利用PCNN的全局耦合特性和脉冲同步特性选择高频系数;最后经Curvelet逆变换得到融合结果。实验结果表明,该方法得到的融合图像在边缘等细节上比传统方法具有更好的视觉效果,在熵、平均梯度、标准差等客观指标上都优于其它方法。

The introduction of the full paper reviews some papers in the open literature and then proposes what we believe to be a better algorithm than previous ones mentioned in the title, which is explained in sections1, 2 and 3, Their core consists of： （1） we perform the fast discrete Curvelet transform of infrared image and visible image one by one and thus obtain the subband coefficients at different scales and in different directions; （2） we fuse the two types of image by conducting the weighted averages of their low-frequency subband eoeflqcients; （3） we select their high frequency coefficients as the inputs to PCNN by using its global coupling characteristics and pulse syn- chronization characteristics and also select regional energy measurements as the connection strengths of PCNN, thus obtaining the fusion results through inverse Curvelet transform. Section 4 eonducts two simulation experiments on our image fusion algorithm to verify its effectiveness; the simulation results, given in Figs. 2 and 3 and Tables 1 and 2, and their analysis show preliminarily that ：. （ 1 ） the images fused with our algorithm have clearer edge and texture details; （2） the objective indexes of our algorithm such as entropy, average gradient, spatial frequency and standard deviation are superior to other existing algorithms.